- TargetColumn
- Specify the name of the input table column that contains the time series data.
- Accumulate
- [Optional] Specify the names of the input table columns to copy to the output table.Tip: To identify change points in the output table, specify the columns that appear in partition_exp and order_by_exp.
- SegmentationMethod
- [Optional] Specify the segmentation method, normal distribution (in each segment, the data is in a normal distribution).
- WindowSize
- [Optional] Specify the size of the sliding window. The ideal window size depends heavily on the data. Experiment with this value.
- ChangePointThreshold
- [Optional] Specify a DOUBLE PRECISION value that the function compares to ln(L 1)−ln(L 0). The definition of Log(L 1) and Log(L 0) are in ChangePointDetection. They are the logarithms of L 1 and L 0.
- OutputType
- [Optional] Specify the output table columns. The schema of the output table is Change-Point Detection Functions Output Table Schema, OutputOption ('CHANGEPOINT') in ChangePointDetection Output.